agentleFS
gbvillarinhogbvillarinho/react-design-skills/copy/SKILL.md

copy

Skill0 starsChanged 33 days agoMITView on GitHub

Writing and reviewing the text of pages and products so it sounds human, specific and useful, not like AI. Use whenever the user needs headlines, hero text, landing page copy, section text, product descriptions, calls to action, about pages, emails or any page text, or wants existing text reviewed or rewritten because it sounds generic, robotic or "AI-written", even if they never say "copy". Works for Portuguese and English. For buttons, errors, empty states and other interface microcopy, use the `ux` skill's writing reference alongside this one.

14 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use scope, prototype and 4 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • YuriNachosSkill

    scope

    Configure claude-scope status line widgets for Claude Code. Use when user wants to customize widgets, change theme, reorder elements, or modify display style.

    Also in 3 projects that use this

  • GoogleSkill

    prototype

    Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

    Also in 4 projects that use this

  • photon-groveSkill

    unslop

    Detects and rewrites generic, overly polished, or AI-sounding prose while preserving meaning. Use when the user asks to "unslop", humanize, de-AI, tighten, or review whether writing sounds generated.

    Also in 3 projects that use this

  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    Also in 4 projects that use this

  • GoogleSkill

    grilling

    Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

    Also in 3 projects that use this

  • MicrosoftSkill

    handoff

    Compact the current conversation into a handoff document for another agent to pick up.

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read it aloud: If it sounds like an advertisement nobody would say to a friend, simplify.

    line 17
  2. Distinctive:

    When asked to review, do not just polish.

    line 30

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Copy

Copy that sounds generated fails for a reason: it is general. It could describe any business, so readers feel nothing and trust nothing. Good copy is specific to this subject, written for one reader, and makes one clear point per line.

Process

  1. Know the facts. Collect what is true and concrete about the subject: what it is, who it is for, what is different, prices, times, places, numbers, materials, process, names. If the brief lacks facts, ask for them or mark clearly what is invented placeholder. Never invent claims that could be mistaken for real ones (awards, years in business, quantities).
  2. Name the reader and the job. Who reads this, what do they want to know, and what should they do next? Write for that person, in their words.
  3. Choose the one point of each piece. A headline, a section or a paragraph says one thing. If you cannot state it in a sentence, the text is not ready.
  4. Draft from the facts, then follow references/page-copy.md for structure by page type.
  5. Review against the AI tells in references/ai-tells.md. Rewrite every line that matches, using a concrete fact instead of a general claim.
  6. Read it aloud. If it sounds like an advertisement nobody would say to a friend, simplify.

Principles

  • Specific beats impressive. "Bread baked at 6 each morning, gone by noon" says more than "authentic, artisanal quality".
  • Say what it is and what happens. Plain nouns and verbs, before any attitude.
  • Don't perform a voice. Match the brand and audience without slogans, wordplay or forced personality.
  • Vary rhythm. Repeated sentence shapes, even clever ones, are the clearest sign of a generated text.
  • Respect the reader's time. Cut anything that does not help them decide or act.
  • Write in the reader's language and variant, with natural idiom, not a translation of an English pattern. Check the locale.

Reviewing existing text

When asked to review, do not just polish. List the lines that match an AI tell or say nothing specific, say why, and offer a rewrite that uses a real fact. If the facts are missing, say what you need to ask the owner.

Working with the other skills

ui decides how the page looks, ux decides structure and behavior, and copy decides what the words say. Copy and layout shape each other: headline length affects the composition, and a design built around a real fact (a schedule, a price list) needs that fact first. For buttons, labels, errors and empty states, follow ux/references/writing.md.

Link text, alt text, labels and error messages are read aloud by screen readers; write them to make sense out of context. See the a11y skill.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.